Context-Semantic Quality Awareness Network for Fine-Grained Visual Categorization

TL;DR

This paper introduces CSQA-Net, a novel weakly supervised network for fine-grained visual categorization that enhances discriminative feature extraction by modeling spatial context and semantic quality, leading to improved performance.

Contribution

The paper proposes a new multi-part, multi-scale cross-attention module and a semantic quality evaluation module to better capture and supervise discriminative features in FGVC.

Findings

CSQA-Net outperforms state-of-the-art methods on four FGVC datasets.

The MPMSCA module effectively models spatial relationships for finer details.

The MLSQE module improves hierarchical semantic supervision.

Abstract

Exploring and mining subtle yet distinctive features between sub-categories with similar appearances is crucial for fine-grained visual categorization (FGVC). However, less effort has been devoted to assessing the quality of extracted visual representations. Intuitively, the network may struggle to capture discriminative features from low-quality samples, which leads to a significant decline in FGVC performance. To tackle this challenge, we propose a weakly supervised Context-Semantic Quality Awareness Network (CSQA-Net) for FGVC. In this network, to model the spatial contextual relationship between rich part descriptors and global semantics for capturing more discriminative details within the object, we design a novel multi-part and multi-scale cross-attention (MPMSCA) module. Before feeding to the MPMSCA module, the part navigator is developed to address the scale confusion problems and accurately identify the local distinctive regions. Furthermore, we propose a generic multi-level semantic quality evaluation module (MLSQE) to progressively supervise and enhance hierarchical semantics from different levels of the backbone network. Finally, context-aware features from MPMSCA and semantically enhanced features from MLSQE are fed into the corresponding quality probing classifiers to evaluate their quality in real-time, thus boosting the discriminability of feature representations. Comprehensive experiments on four popular and highly competitive FGVC datasets demonstrate the superiority of the proposed CSQA-Net in comparison with the state-of-the-art methods.